Food source and trophic position analysis of spiders living in northern wetlands

BIOS 35502: Practicum in Field Biology

UNDERC-East

Adriana Cintrón

Advisor: Keith O’Connor

2017

Cintrón 2

ABSTRACT

Wetland ecosystems are important ecological features that contain complex food webs due to their terrestrial and aquatic inputs. Feeding and diet habits of spiders in different types of wetlands is poorly understood, but is an essential subject to understand different interactions and community structure of organisms within these ecosystems. The δ13C and δ15N isotopic signatures have been found to be effective tools to better understand and interpret species’ diets and trophic levels. Spiders are generalist predators, and are often cannibalistic, causing their isotopic signatures to be further complicated. Given the low insect productivity in wetlands during the dry season, spiders are expected to occupy a higher trophic level. We analyzed the nitrogen and carbon isotopic signatures of spiders and vegetation during the dry season from three different wetland locations: two vernal pools and one bog. We didn’t find any significant differences for isotopic signature between spider families or between sites. The average δ13C value for spiders and vegetation were – 25.620 ± 1.095 and -28.758 ± 1.385 respectively. These values indicate that C3 vegetation serves as the basal carbon source and that spiders rely on terrestrial food source. The average δ15N value for spiders and vegetation were 5.486 ± 0.940 and -1.790 ± 1.092, respectively. There was approximately a 7‰ 15N enrichment in spiders. Using the literature value of approximately 3 to 4‰ 15N increase per trophic level, the found 7‰ 15N enrichment suggests spiders occupy two levels up in the trophic chain. Our study provides evidence for utilizing spider family isotopic signatures to better understand food webs in wetland ecosystems.

Keywords: Spiders; Isotopic composition; Generalist predators; Wetlands; Trophic interactions; Food source; Stable isotope analysis

INTRODUCTION

Spiders are dominant predators that consume a large variety of prey and engage in

intraguild predation (Mestre et al. 2013). Their great abundance, biological functionality, and

diversity have enabled them to inhabit a vast variety of ecosystems and living environments.

They also have many important ecological roles, including serving as a source of nutrient input in

ecosystems (Hodkinson et al. 2001). Spiders provide a mechanism for the natural deposition of

dead insects, a ready source of nutrients that optimizes nutrient retention and

distribution (Hodkinson et al. 2001). Despite this, information pertaining to spiders is sparse. We

Cintrón 3

still lack knowledge about the biology and feeding habits of spiders in their natural environment

(Cardoso et al. 2011; Greenstone 1999; Mestre et al. 2013).

Diets of animals have been studied and used to better understand their relationship and

interactions with their environment and other organisms, especially for trophic level

studies (Mestre et al. 2013; Hyodo 2015; Iakovlev et al. 2017). Various studies (Akamatsu et al.

2004; Paetzold et al. 2005) have used the isotopic composition of riparian spiders to study their

food source. Riparian spiders interact with both aquatic and terrestrial ecosystems, and as

such their diets consists of a variety of food sources, which is reflected in their isotopic

composition. Aquatic food sources have a different isotopic composition from terrestrial food

sources (Hyodo 2015). However, these studies focus on spiders living in areas adjacent to large

bodies of freshwater, and not in wetland ecosystems.

Wetlands have different ecological and biological properties from rivers or streams. They

also sustain a different variety of terrestrial and aquatic organisms. The smallest and often the

most ephemeral of the wetlands are temporary or vernal pools that fill from seasonal rains or snow

melt (Batzer & Wissinger 1996). These pools can flood and dry repeatedly, fill for a few months,

or remain flooded for much of the year. Insect productivity can be very high in temporary pools,

possibly because decomposition during waterless stages enhances detrital food quality (Batzer &

Wissinger 1996). Peat develops in wetlands where decomposition does not keep pace with plant

production, such as bogs (Batzer & Wissinger 1996). Peat accumulation cause peatland surfaces

to gradually rise above the water table and thus many of the insects found in these habitats are of

terrestrial origin (Batzer & Wissinger 1996), although aquatic species can occur in the water

around peatland vegetation itself (Batzer & Wissinger 1996). In these fishless habitats, predaceous

insects can become very abundant, and they are often apex aquatic predators (Batzer & Wissinger

Cintrón 4

1996). Spiders are dominant generalist predators that consume a great variety of prey, engage

in intraguild predation, and reduce herbivore numbers in terrestrial habitats (Hooks et al 2006).

Therefore, they are an important component in these ecosystem interactions.

This study focuses on feeding habits, trophic position, and the diet of spiders living in

northern wetlands, specifically bogs and vernal pools. The analysis of the natural variations in

stable isotope ratios in animal tissue is a powerful tool to reveal feeding preferences of generalist

predators such as spiders (Oelbermann & Scheu 2001). This approach is based on the fact that

C and N isotopic signatures of body tissues of consumers reflect those of its diets (Hyodo 2015).

Stable carbon isotope ratios are also used increasingly to examine the trophic flow across terrestrial

and freshwater ecosystems (Kato et al. 2004). Stable isotope analysis enables characterization of

trophic relationships between organisms because it tracks the energy flow in food webs (Mestre et

al. 2013). The δ13C isotopic signature has been proven to indicate the basal carbon source of a

species (Hyodo 2015). The δ15N of a species indicates the average number of trophic transfers

between the species and the base of the food web with enrichment values ranging 3 to 5‰ per

trophic level (Mestre et al. 2013).

In this study, we seek to understand how multiple spider families use the available food

sources from terrestrial and aquatic ecosystems. More specifically, we seek to study food source

preference, aquatic or terrestrial, of spiders using their isotopic signatures to better understand

trophic interactions in wetland ecosystems during the dry season. The questions being addressed

are 1) are terrestrially or aquatically derived food sources the larger contributor to spiders’ diet and

2) how can we describe their trophic position in northern wetland ecosystems, specifically vernal

pools (sites 1 & 3) and a bog (site 2) during the dry season? To determine this, spiders and

vegetation from each site were collected and run on an elemental analyzer coupled to an Isotope

Cintrón 5

ratio mass spectrometer (EA-IRMS) for bulk analyses of stable C and N isotopes. Carbon stable

isotopes are useful for tracing basal carbon sources, whereas nitrogen stable isotopes are greater

indicators of trophic level (Newton 2016). We hypothesized terrestrial food sources will be a

greater contributor to spiders’ diet in northern wetland ecosystems than aquatic food sources and

that spiders are predominantly two to three trophic transfers away (7-12‰ 15N enrichment) from

the basal carbon source, since they serve as generalist dominant predators in the ecosystem (Lang

et al. 1999) and are also cannibalistic.

MATERIALS AND METHODS

Field sampling

Sweep nets and pitfall traps were used to collect spiders at three different locations

(Figure 1) at the University of Notre Dame Environmental Research Center (UNDERC) in

Wisconsin, USA. Sampling locations included three wetland areas, including two vernal ponds

and one bog. The pitfall traps consisted of five buckets that were dug into the ground at

approximately 2 meters away from each other and netting placed between them in a cross

arrangement (Figure 2). Sampling periods took place in the morning and evening for three weeks

throughout the summer (May 25th - June 3rd, June 11th - 17th and July 9th -12th). Spiders were

collected using labeled ethanol rinsed glass jars and vials, and were carried back to the lab.

Dominant vegetation samples were also collected for basal carbon source analysis on 17/06/13.

Plants were selected based on their commonality between all three sites, as well as visible

evidence of insect interactions, such as bite marks, larvae on leaves, etc. Site assessment for the

three locations was done during the last research week. Notes were taken on site vegetation

composition. Soil moisture and pH were measured using a Kelway Soil Tester sensor.

Cintrón 6

Sample processing

Samples were allowed to stay in the jars overnight or for 12 hours to give enough

time for complete food digestion. Samples were then frozen in a -20 ̊ C freezer for at least 8

hours. A Leica EZ4 HD photographic dissecting microscope was used to identify spiders down

to family level. Photos were taken using Leica Application Suite software. After identification,

samples were placed in a drying oven at 50 ̊ C for 12 hours, or until dry. Samples were ground to

a fine powder using an ethanol rinsed pestle and mortar. Each powdered sample was transferred

into a labeled penny-envelope and sealed for transport to Notre Dame’s stable isotope facility

located in Center for Environmental Science and Technology (CEST).

Each sample was weighed to around 0.8 mg on a microbalance and placed into 4 x 6 mm

pressed tin capsules. This process was repeated for each sample. The stable carbon and nitrogen

isotope ratios were found using an elemental analyzer coupled to a Delta V Isotope ratio mass

spectrometer (EA-IRMS). Stable C and N isotope ratios are expressed as:

δ13C or δ15N = [(Rsample/Rstandard) - 1] X 1000,

where R is 13C/12C or 15N/14N for δ13C or δ15N, respectively. The standard for C is the Peedee

Belemnite (PDB), and for N the standard is atmospheric N2.

Data analysis

Samples were analyzed for δ13C and δ15N using an elemental analyzer coupled with an

isotope ratio mass spectrometer (EA-IRMS). In-house standards of sorghum flour, protein

powder, and peach leaves standards were used to correct isotopic values. An acetanilide standard

was used to calculate the %C and %N using the following equations:

Cintrón 7

a) %C = 71.09 x std. amount (mg) std. peak area 44

X sample peak area 44 sample amount (mg)

b) %N = 10.36 x std. amount (mg) std. peak area 28

X sample peak area 28 sample amount (mg)

All four standards were run at the beginning and the end, with three run in the middle to

ensure no drifting occurred during the run. Only samples from runs with an R2 value greater than

0.99 were utilized for analysis. All the samples were run in duplicate to ensure reproducibility.

Statistical analysis

Measured and expected isotope signatures for the standards were analyzed using

regressions to obtain linear equation to be used for correcting data. Three spider samples were

dropped when running statistical tests and descriptive statistics since they fell outside the 2

standard deviation and 95% of the data was still represented. Since data was not normally

distributed, Kruskal-Wallis test was used to test for any C:N differences between families and

sites. All statistical tests were run using the program Systat, version 10.0.

RESULTS

Data was not normally distributed and could not be transformed, so non-parametric

Kruskal-Wallis testing was employed. The first Kruskal-Wallis tested for any difference in C:N

ratio between families. No significant difference in C:N ratios was detected between spider

families using a Kruskal-Wallis test (test statistic= 7.621, df=4, p-value=0.1062).

The second Kruskal-Wallis tested for C:N ratio differences between type of site; site 1

being vernal pools and site 2 being one bog. No significant difference in C:N ratios was detected

between types of sites using a Kruskal-Wallis test (test statistic= 2.256, df=2, p-value=0.324).

Thus, there was not a significant difference in C:N ratio depending on type of site.

Cintrón 8

The averages for δ13C and δ15N for spiders were -25.620 ± 1.095 and 5.486 ± 0.940, and

for plants they were -28.758 ± 1.385 and -1.790 ± 1.092, respectively. The δ13C values for the

plants and spiders were similar. For δ15N, there was approximately an increase of 7‰. The C:N

ratio average for plants was 3.310 ± 0.852. For spiders, it was 7.076 ± 0.225 (Figure 5). The

average C:N ratio by type of site for both spiders and plants was also calculated. For site 1

(vernal pool), spiders had an average C:N ratio of 7.091 ± 0.078 , and plants had an average of

2.915 ± 0.059. For site 2 (bog), spiders had an average C:N ratio of 7.032 ± 0.410, and plants

had an average of 3.569 ± 0.634. Generally, spiders have a larger C:N ratio than plants in both

types of wetlands studied (Figure 6). In addition, values for C and N isotope signature for spider

families showed no significant variation (Table 1).

DISCUSSION

There was not a significant difference for C isotope signature between different spider

families (Figure 7). This suggests that basal carbon source does not significantly vary between

spider families. However, different abiotic and biotic factors could have influenced these results.

Since the study was carried during summertime, drought and water evaporation rates dried the

sites. In addition, aquatic insect composition would have been reduced drastically, as conditions

were not favorable anymore and water was scarce. A previous study (Kato et al. 2004) carried in

Japan found that δ13C for linyphiid spiders showed a preference for aquatic prey, except in June-

July; that is, the isotopic composition started to resemble that of terrestrial prey during

summertime. In contrast, araneid spiders relied on terrestrial prey throughout the whole study

period. This suggests different groups of spiders can have different diet preference, but prey can

also change seasonally. In addition, it is possible that aquatic insect productivity was low for that

particular vernal pool, and so there would have been less aquatic food resources. If wetlands are

Cintrón 9

flooded infrequently or for short durations, their invertebrate productivities may be low (Batzer &

Wissinger 1996). Petit (2017) explained that the capacity of some wetlands in Australia to provide

refugia for aquatic biota during the dry season is likely to be highly dependent on either localized

runoff or ground water inputs. It is possible that the same criteria can also be applied for some

northern wetlands.

Additionally, there also wasn’t a significant difference in δ15N values between families.

Spiders are mostly known as predators, and so are expected to prey on similar functional

organisms. However, this is not always the case. A previous study showed that jumping spiders

have δ13C values similar to acacia plant species, and δ15N values higher by 2.9‰ than the Beltian

food bodies of acacia species (Hyodo 2015). However, this study was carried in grasslands, and

nutritious plants were abundant.

The values for δ13C shows us the basal carbon source, whereas δ15N indicates trophic level

(Hyodo 2015; Perkins et al. 2014; Michener & Kaufman 2008). The similarity of N isotopic

signatures values between spiders indicates that they tend to occupy similar trophic levels. This

suggests that spiders tend to prey on similar functional organisms, at least during the period this

study was held. On the other hand, the similarity of C isotopic signatures suggests that they belong

to similar food chains, since their C isotope signature reflects that of the same C3 vegetation and

thus their primary source of carbon. The C isotopic signature helps us differentiate between carbon

sources that greatly differ in their δ13C values, such as C3 or C4 photosynthetic pathways or aquatic

and terrestrial sources (Michener & Kaufman 2008). For the collected plants, the average C

isotopic signature was -28.758 ± 1.385. C3 plants have δ13C values averaging about -26.5‰, while

values for C4 species average about -10 to -12.5‰ (Hyodo 2015; Michener & Kaufman 2008).

The C isotopic signature of the plants resemble that of the plants that undergo C3 photosynthetic

Cintrón 10

pathways. This makes sense since most northern deciduous trees and conifers use C3

photosynthetic pathway.

The δ13C average for spiders was -25.620 ± 1.095, which indicates that the primary carbon

source is the C3 plants. Overall, there is a small (0.5-1.5‰) enrichment in δ13C in the organism

relative to its diet (Michener & Kaufman 2008). Variations in δ13C could be due to the variation

of isotopic composition of individual tissues within organisms, which can reflect differences in

amino acid and lipid composition and concentration, variation among turnover time for different

tissues and different rates of tissue turnover (Michener & Kaufman 2008). Moreover, cases where

δ13C depletion occurs are not unknown. In a previous study, larvae of some aquatic insects such

as Helode ssp. in detritus of stream backwater pools, and the adults collected on the stream bank,

were both highly depleted in δ13C (−67‰) (Kohzuet al. 2004). This methane-derived C is

suggested to support terrestrial predators such as spiders (Jones & Grey 2011).

The δ13C values of aquatic plants usually lie in an intermediate position between the values

of terrestrial C3 and C4 plants, depending on their carbon sources. Aquatic fauna will usually have

carbon isotopic values that are less negative than those of terrestrial animals feeding on C3-based

foods and more negative than those feeding on C4-based foods (Michener & Kaufman 2008).

However, freshwater aquatic food webs can appear to have C3-like carbon isotopic compositions.

The stable N isotopic signatures of plants are known to be influenced by various factors

such as climate and the successional stage of the ecosystem (Hyodo 2015). However, the three

locations had characteristics and vegetation of an early successional stage ecosystem, so this

wouldn’t be a determining factor. Nitrogen enrichment increases 3-4 ‰ with trophic level (Hyodo

2015; Perkins et al. 2014). The δ15N values of most modern plants are between 0 and 5‰.

Therefore, the nitrogen isotopic values of populations that rely on terrestrial animal protein in their

Cintrón 11

diet (two trophic transfers), would average from around 6-9‰. The average δ15N value for plants

collected was -1.980 ‰. This may have been due to different climatic and physiological factors.

In contrast, the average for spiders was 5.486‰. If we compare the values for plants and spiders,

we see that there was approximately an increase of 7‰. This indicates that spiders occupy two

positions up in the trophic chain. In addition, this also suggests that spiders in these wetland areas

rely more heavily on terrestrial animal protein. The nitrogen levels of a consumer will increase

more than 6-9‰ when relying more heavily on aquatic resources (Schoeninger & DeNiro, 1984).

As expected, spiders have a larger C:N ratio than plants (Figure 6). This could possibly

reflect the spiders’ usage of carbon for more structural features. This also suggests that plants are,

as expected, a better quality food source than spiders, indicated by the lower C:N ratio (Petit 2017).

Overall, stable isotopes were found to be a useful tool for better understanding the food web of

three wetlands in northern Wisconsin. The stable isotopic ratios for C and N were found to be

useful indicators of basal plant sources and trophic positions. Improvements, such as having more

replicates and identifying spiders’ species could reflect the existence of marked trophic differences

between spider species belonging to the same family (Mestre et al 2013). This could allow for a

complete construction of food webs and would help to better understand interactions between

organisms in wetland ecosystems.

Cintrón 12

ACKNOWLEDGEMENTS

I would like to thank my mentor, Keith O’Connor, for guiding me and providing feedback and

advice while conducting this study, as well as preparing and running samples on CEST’s EA-

IRMS. I would also like to thank Hannah Legatzke, Claire Goodfellow and Bethany Hoster for

their help in analyzing data and writing this paper. Thanks to Renee Dollard for assisting me in

conducting this study and collecting samples for this experiment. I want to thank Mariana Cruz for

aiding me in the identification of spiders. I would like to thank Patrick Larson and Michael Cramer

for their help in sending samples to the University of Notre Dame so they could be analyzed. Gary

Belovsky and Michael Cramer for directing this course, the Bernard J. Hank Family Endowment

and José Enrique Fernández, for the financial support which made it all possible. Lastly, I thank

the University of Notre Dame for making this study and my participation in this course possible.

Cintrón 13

REFERENCES

Achim Paetzold, Carsten J. Schubert, and Klement Tockner; Aquatic Terrestrial Linkages Along a Braided-River: Riparian Arthropods Feeding on Aquatic Insects. Ecosystems 2005 8: 748–759

Akamatsu, F., Toda, H. and Okino, T. Food source of riparian spiders analyzed by using stable isotope ratios. Ecological Research 20014; 19 (6): 655-662

Batzer, D. and Wissinger, S. Ecology of Insect Communities in Nontidal Wetlands. Annu Rev.

Entomol. 1996; 41:75-100 Hyodo, F. Use of stable carbon and nitrogen isotopes in insecttrophic ecology. Entomological

Science 2015;18,295–312

Iakovlev, I. et al. Trophic position and seasonal changes in the diet of thered wood ant Formica aquilonia as indicated by stable isotope analysis. Ecological Entomology 2017; 42: 263–272

Ian D. Hodkinson, Stephen J. Coulson, Joanna Harrison, Nigel R. Webb. What a wonderful web they weave: spiders, nutrient capture and early ecosystem development in the high Arctic – some counter-intuitive ideas on community assembly. Oikos 2001: 95(2): 185–360

Kato, C., Iwata, T. and Wada, E. Prey use by web-building spiders: stable isotope analyses of trophic flow at a forest-stream ecotone. Ecological Research 2004; 19(6): 633-643

Keith A. Hobson, Harold E. Welch; Determination of trophic relationships within a high Arctic marine food web using carbon-13 and nitrogen-15 analysis. Marine Ecology Progress Series 1992; 84(1): 9-18

Lang, A., Filser, J. and Henschel J. Predation by ground beetles and wolf spiders on herbivorous insects in a maize crop. Agriculture, Ecosystems and Environment 1999; 72:189-199

Margaret J Schoeninge, Michael J. DeNiro; Nitrogen and carbon isotopic composition of bone

collagen from marine and terrestrial animals. Elsevier 1984; 48 (4): 625-639

Cintrón 14

Meier-Augenstein, W. and Kemp, H. F. 2012. Stable Isotope Analysis: General Principles and Limitations. Wiley Encyclopedia of Forensic Science.

Mestre, L et al. Trophic structure of the spider community of a Mediterranean citrus grove: A stable isotope analysis. Basic and Applied Ecology 2013; 14 (413-422)

Michener, L. K. and Kaufman. R. 2007. Stable isotope ratios as tracers in marine food webs: and update. Pages 238-282 Stable Isotopes and Ecology in Environmental Science. Blackwell Publishing, Oxford, UK.

N. E. Pettit, D. M. Warfe, P.G. Close, B. J. Pusey, R. Dobbs, C. Davies, D. Valdez, C and P. M.

Davies. Carbon sources for aquatic food webs of riverine and lacustrine tropical waterholes with variable groundwater influence Marine and Freshwater Research, 2017,68, 442–451

Newton, J. (2016). Stable Isotopes as Tools in Ecological Research. eLS.

Perkins, M et al. (2014) Application of Nitrogen and Carbon Stable Isotopes (δ15N and δ13C) to Quantify Food Chain Length and Trophic Structure.

Oelbermann K, Scheu S (2002) Stable isotope enrichment (δ15N and δ13C) in a generalist predator (Pardosa lugubris, Araneae: Lycosidae): effects of prey quality. Oecologia 130: 337–344.

Schoeninger, M. J. and M. J. DeNiro. "Nitrogen and Carbon Isotopic Composition of Bone Collagen from Marine and Terrestrial Animals", Geochimica et Cosmochimica Acta 1984; 48:625-639.

Cintrón 15

GRAPHS AND FIGURES

Family Average C isotope signature

Average N isotope signature

Lycosidae -24.928±0.736 5.144±0.947 Thomisidae -23.816±0.842 5.309±0.337 Pisuridae -24.476±0.057 6.106±0.041 Araneidae -26.923±0.225 6.951±0.0710 Agelenidae -26.695±0.034 6.318±0.008

Table 1. Average C and N isotope signature for each isotope family. There is not significant variation among families.

Figure 1: Red markings represent sampling locations

Cintrón 16

Figure 2: Pitfall trap arrangement

Figure 3: N and C isotopic signatures for spider families

0

1

2

3

4

5

6

7

8

-30 -20 -10 0

δ15

N (‰

)

δ 13C (‰)

Lycosidae

Pisuridae

Araneidae

Thomisidae

Agelenidae

2 m

Cintrón 17

Figure 4: N and C isotopic signatures for spider families and sampled vegetation. There is approximately a 7 permil increase in nitrogen isotope signature. C isotope signature values are consistent and similar, indicating C3 vegetation as basal carbon source.

-4

-2

0

2

4

6

8

-40 -30 -20 -10 0

δ15

N (‰

)

δ 13C (‰)

Lycosidae

Pisauridae

Araneidae

Thomisidae

Agelenidae

Vegetation

Cintrón 18

Figure 5: Average C:N ratio for different spider families. There is not a significance difference for C:N ratio among families.

6.6

6.7

6.8

6.9

7

7.1

7.2

7.3

Lycosidae Thomisidae Pisauridae Araneidae Agelenidae

Aver

age

C:N

ratio

Family

Average C:N ratio

Cintrón 19

Figure 6: Average C:N ratio for different spider families by type of site. There is not a significant difference for C:N ratio among families depending on type of site. This suggests type Site 1 refers to vernal pool and site 2 to a bog.

0

1

2

3

4

5

6

7

8

Site 1 Site 2

Aver

age

C:N

ratio

Type of site

Average C:N ratio

Spiders Plants

Cintrón 20

Figure 7: Average C and N isotopic signatures for different spider families. There is not a significant difference in C and N isotopic signatures among families.

-30

-25

-20

-15

-10

-5

0

5

10

Lycosidae Thomisidae Pisuridae Araneidae Agelenidae

Aver

age

C an

d N

isot

opic

sig

natu

res

Family

Average C and N isotopic signatures

Average C isotope signature Average N isotope signature

Cintrón 21

Figure 8: Lycosidae

Figure 9: Pisauridae